Defining mediators and mechanisms which favor the early stages of the colon cancer such as formation of adenoma, is integral to improved surveillance and prevention of colon cancer. Our recent studies demonstrate strong correlations between reduced apoptosis in the normal intestinal mucosa and risk of precancerous adenomatous lesions in humans. Preliminary and published evidence indicates elevated plasma insulin and increased levels or bioavailability of circulating or locally expressed insulin-like growth factors (IGF-I or IGF-II) as significant risk factors for reduced apoptosis and adenoma in the colon. Common mediators of insulin and IGF action on apoptosis or adenoma growth may represent particularly powerful biomarkers of adenoma risk. Insulin receptor substrate 1 (IRS-I) is a common mediator of the anti-apoptotic actions of IGF-I, IGF-II and insulin. Preliminary data in mouse models demonstrate that IRS-1 is required for anti-apoptotic actions of IGF-I in the normal intestine in vivo and that loss of just 1 copy of the IRS-1 gene increases apoptosis and reduces adenoma formation in the APC/Min-+model of intestinal polyposis. Recent studies link polymorphisms in the human IRS-1 gene to colon cancer risk. Together, this data supports our central hypothesis that levels of expressed IRS-1 in normal human colonic mucosa predict the levels of apoptosis and adenoma risk by mediating the anti-apoptotic actions of insulin and the IGFs. Our studies will use existing samples from a large case-control study of colorectal adenomas to test our hypothesis. Data on apoptosis, plasma insulin, IGFs and plasma or colonic levels of insulin-like growth factor binding proteins (IGFBPs) have already been collected in this same study population. Aim 1: will test the hypothesis that a Gly972Arg polymorphisms in the IRS-I gene, and/or levels of expressed IRS-1 mRNA in the normal colonic mucosa predict adenoma risk and levels of apoptosis. Genotyping will be performed by PCR based allelic discrimination and IRS-1 mRNA will be measured by real time PCR. Primary analyses will test for associations between IRS-1 genotype or expression levels and adenoma or apoptosis. Secondary analysis, we will test if there are interactions between IRS-1 genotype or IRS-1 expression with insulin/IGFs or IGFBPs in predicting apoptosis and adenoma risk. Aim 2 will test the hypothesis that loss of imprinting (LOI) for the IGF-II gene increases local IGF-II expression in colon and increases risk of adenoma or low apoptosis. Subjects heterozygous for 820G/A and 266C/T polymorphisms in the IGF-II gene will be identified. Allelic and total expression levels for IGF-II will be assessed by real time PCR. LOI for IGF-II and IGF-II expression will be tested for associations with adenoma or apoptosis and for interactions with IRS-1 genotype or expression levels in predicting adenoma risk or apoptosis.